Complex Parameterization Research Articles

Comparison of adaptive techniques to predict crop yield response under varying soil and land management conditions

Decision-making processes in agriculture often require reliable crop response models to assess the impact of specific land management. While process-based models are often preferred over empirical ones in current modelling communities, empirical crop growth models can play an important role in identifying the hidden structure of crop growth processes relating to a wide range of land management options. This study investigates the potential of predicting crop yield responses under varying soil and land management conditions by applying three different adaptive techniques: general linear models (GLMs), artificial neural networks (ANNs), and regression trees (RTs). The crop yield data used in this research consist of 720 maize yield indices from 11 different land management trials in southern Uganda. GLM showed the poorest results in terms of modelling accuracy, prediction accuracy, and model uncertainty, which might suggest its inability to model the non-linear causal relationships present in complex soil–land and crop-management interactions. The other two non-parametric adaptive models show significantly higher prediction accuracy than GLM. RT is the most robust technique for predicting crop yield at the study site. ANN is also a promising tool for predicting crop yield and offers insight into the causal relationships through the use of sensitivity analyses, but the complex parameterization and optimum model structure require further attention. The three adaptive techniques compared in this research showed different advantages and disadvantages. When these methods are used together, valuable information can be provided on crop responses, and more reliable crop growth models may result.

Topographic steering of dense currents with application to submarine canyons

The influence of a submarine canyon on a dense bottom current flowing along the continental shelf is considered. It is shown that the front of the current moves forward with the Nof velocity, establishing a stationary plume parallel to the depth contours behind it. The stationary flow has a frictional transport directed downhill, perpendicular to the depth contours. This transport converges inside topographic depressions. Consequently, the heavy water accumulates in such areas and the current may change direction and flow inside the depression rather than along the depth contours. Submarine canyons can influence the along-shelf flow in this manner. A canyon that intersects the continental slope has the ability to channel dense water down to the deep sea. The rate at which water is drained from the along-shelf flow is determined from the canyon's geometry and the magnitude of bottom friction. If the along-shore transport upstream of the canyon exceeds the capacity of the canyon, the plume divides into two branches. A portion of the flow is then steered down into the canyon while the rest continues on the downstream side of it. The distribution of dense water inside the canyon is obtained analytically for a simplified description of bottom friction, as well as numerically for more complex friction parameterizations. The results are remarkably similar. This indicates that the downward flow is controlled by the topography and the magnitude of bottom friction rather than the choice of how friction is parameterized.

A GIS‐supported model for the simulation of the spatial structure of wildland fire, Cass Basin, New Zealand

Summary 1. The behaviour of wildland fire in spatially heterogeneous landscapes was simulated using a model (PYROCART) that integrates the Rothermel fire spread model and a geographic information system (GIS). 2. The principal aims of the research were to test the applicability of overseas fire behaviour models to New Zealand ecosystems, and to assess the applicability of GIS to fire spread prediction. 3. The model was validated using an uncontrolled fire that occurred in the Cass Basin, South Island, New Zealand in May 1995. This fire burnt 580 ha across a complex vegetation mosaic comprising shrubland, stands of Nothofagus solandri var. cliffortioides, bog and tussockland. 4. The overall predictive accuracy of the model was estimated to be 80%. Prediction accuracies within different fuel types and slope angles are also presented. Fuel type and slope appeared to be the dominant influences on fire spread. No trends in prediction accuracy by wind speed or direction were apparent. The predicted burnt area and the real burnt area had a similar overall shape. It was found, however, that at high wind speeds the model tended to over‐predict rates of fire spread in some directions. 5. The PYROCART model shows potential as a land management tool, especially for the testing of hypotheses concerning land management strategies. However, due to the complex input data and parameterization techniques it requires, it is less suitable for in situ fire management.

Measurement of free radicals OH and HO2 in Los Angeles smog

Atmospheric free radicals hydroxyl and hydroperoxyl (OH and HO2, collectively HOx) are the catalysts that cause secondary or photochemical air pollution. Chemical mechanisms for oxidant and acid formation, on which expensive air pollution control strategies are based, must accurately predict these radical concentrations. We have used the fluorescence assay with gas expansion (FAGE) technique to carry out the first simultaneous, in situ measurements of these two radicals in highly polluted air during the Los Angeles Free Radical Experiment. A complete suite of ancillary measurements was also made, including speciated hydrocarbons, carbon monoxide, aldehydes, nitric oxide, nitrogen dioxide, and ozone along with meteorological parameters. Using this suite of measurements, we tested the ability of a lumped chemical mechanism to accurately predict radical concentrations in polluted air. Comparison of model predictions with measured radical concentrations revealed generally good agreement for OH early and late in the day, including the early evening hours, when OH persisted at low concentrations after dark. During midday, however, modeled [OH] was high by about 50%. Agreement for HO2 was quite good in the early morning hours, but model‐calculated HO2 concentrations were significantly too high during midday. When we used our measured HO2 concentrations as model input, agreement between calculated and measured OH concentrations was improved. It seems likely that (1) the model's HOx sources are too large, (2) there are unaccounted HOx loss processes in Los Angeles air, and/or (3) the complex parameterization of RO2/HO2 radical chemistry in the reaction mechanism does not adequately describe the behavior of these radicals in the Los Angeles atmosphere.

Energetics of the 6000-yr BP Atmospheric Circulation in Boreal Summer, from Large-Scale to Monsoon Areas: A Study with Two Versions of the LMD AGCM

Abstract The mechanisms of northern summer monsoon changes at 6 kyr BP and their dependency to model parameterizations is investigated using two versions of the atmospheric general circulation model developed by the Laboratoire de Meteorologie Dynamique, CNRS, Paris (LMD), within the Paleoclimate Modeling Intercomparison Project (PMIP). These two versions differ in their horizontal resolution and in their treatment of some surface processes: one model has a low horizontal resolution and rather simple surface treatments; the other model has a better resolution and uses more complex surface parameterizations. The authors focus on the energy budgets, which are useful tools to understand the impact of the model parameterizations. As a response to mid-Holocene insolation change, the June–September monsoon is enhanced over northern Africa, northern India, and the western Pacific, both in terms of precipitation and low-level convergence changes;convection decreases over Central America. These changes, which are ...

On a Distinguished Quality of the Im B kq/Re B kq Parameter Ratios in the Multipole Expansions of the Crystal Field Potentials

It is demonstrated that for crystal field potentials having complex parameterizations characterized by low symmetry and large number of independent parameters, the formulas for the Im B k q /Re B k q ratios, i.e., for the parameters of the same k and | q|, correspond to a more general superposition model requiring the local metal-ligand symmetry to be ν instead of C ∞ν if the scaling factors alone are treated phenomenologically. A formal expression revealing the mutual interplay of both the distance and the nonaxial scaling factors has been obtained which allows the ratios Im B k q /Re B k q to be estimated more adequately. The generalization is substantial indeed; the corresponding number of independent intrinsic parameters of kth degree rises from one to k + 1. The distinguished quality of these ratios allows them to be treated in a distinguished way during the fitting of the potentials to experimental data. The proper fitting is then easier, since the number of independent parameters can be reduced.

A Comparison of Simple and Complex Treatments of Surface Hydrology and Thermodynamics Suitable for Mesoscale Atmospheric Models

Abstract Alternative treatments of the hydrologic and thermodynamic processes at the earth's surface within a mesoscale model are discussed in this study. Specifically, the question of under what circumstances it is necessary to use a complex surface parameterization scheme as opposed to simpler ones is addressed. Three versions of a one-dimensional planetary boundary layer model were employed, where the primary differences among them are in their surface modules. One uses a simple treatment of the surface characteristics (time independent). In another, the surface processes are represented by a complex surface physics-soil hydrology scheme, while the third one is similar to the first one but the moisture-availability parameter has a specified temporal variation during and after a precipitation event. Several numerical simulations were performed. They showed that the models’ solutions differ the most when the vegetation cover and the surface net radiative flux are large, and the soil-water content cannot ...

Absolute positioning using Doris tracking of the SPOT‐2 satellite

The French DORIS doppler satellite tracking/positioning system has been operational on the SPOT‐2 satellite since 1990. We have analyzed 80 days of data from 42 ground beacons and determined three dimensional site coordinates which agree with a set of SLR/VLBI positions at the 20–30 millimeter level in each coordinate after removal of common rotations, translations, and a scale difference. The accuracy of the vertical component is comparable to that of the horizontal components, indicating that residual troposphere error is not a limiting factor. The translation parameters indicate that the DORIS network realizes a geocentric frame to about 50 mm in each component. The considerable amount of data provided by the nearly global, all‐weather DORIS network allowed this complex parameterization required to reduce the unmodelled forces acting on the low Earth satellite. The positioning results obtained from tracking using essentially the same ground network and the higher altitude TOPEX/POSEIDON spacecraft should be superior to those obtained for SPOT‐2.

A simple parameterization of sub-grid scale open water for climate models

The effects of small fractions ( < 30%) of open water covering a grid element are currently neglected even in atmospheric general circulation models (AGCMs) which incorporate complex land surface parameterization schemes. Here, a method for simulating sub-grid scale open water is proposed which permits any existing land surface model to be modified to account for open water. This new parameterization is tested as an addition to an advanced land surface scheme and, as expected, is shown to produce general increases in the surface latent heat flux at the expense of the surface sensible heat flux. Small changes in temperature are associated with this change in the partitioning of available energy which is driven by an increase in the wetness of the grid element. The sensitivity of the land surface to increasing amounts of open water is dependent upon the type of vegetation represented. Dense vegetation (with a high leaf area index) is shown to complicate the apparently simple model sensitivity and indicates that more advanced methods of incorporating open water into AGCMs need to be considered and compared against the parameterization suggested here. However, the sensitivity of one land surface model to incorporating open water is large enough to warrant consideration of its incorporation into climate models.

Estimating Ability With the Wrong Model

No model is ever a perfect reflection of the data it is to summarize. There are always errors of fit. This is as true with modern item response theory (IRT) as with all other models. It is important to know to what extent the accuracy of measurement made with these models is influenced by misfit and what can be done to minimize the inaccuracy. First, a detailed general model was fit to data to provide the framework for a realistic simulation structure. Then three of the most commonly used IRT models were fit in this simulation. A variety of robust estimators of ability were used and the accuracy and efficiency of each estimator was determined. With short tests, a simple model coupled with a robust estimator seemed to be the methodology of choice for describing the data. As test length increased, so too did the benefits of utilizing a more complex parameterization. An unexpected finding was that coupling robust estimators with a Bayesian prior yielded substantial shrinkage. Future work on ability estimation, especially for practical applications of adaptive testing, is required to “unshrink” ability estimates.

Direct measurement of the mantle attenuation operator from broadband P and S Waveforms

Displacement and velocity records of P and SH body waves with broadband frequency content from 0.01 to 5.0 Hz are obtained either directly or through multichannel deconvolution of earthquake waveforms from digitally recording stations of the Global Digital Seismographic Network, the Regional Seismic Test Network, and the Graefenberg array. Source parameters of a deep focus earthquake are determined by modeling P waveforms with a dynamically realistic source model and a model of Qα that is allowed to be frequency‐dependent. Because the phase behavior predicted by intrinsic attenuation often contrasts sharply with that predicted by realistic rupture models, there are strong constraints on source and attenuation operators. Using the rupture model derived from the P waves and neglecting the effects of scattering, a search is made for the minimum phase operators required to obtain matches to the observed S waves. The variation in the inferred minimum phase operator between S and ScS phases and its behavior with increasing distance and frequency are used to estimate a depth‐ and frequency‐dependent model of intrinsic Qβ. Strong constraints on the frequency dependence of Qβ are given by fine details in the shape and dispersion of the broadband displacements of the S waves. Rise times are most sensitive to the higher end of the frequency band (0.2–1.0 Hz), while the pulse width and tail are sensitive to the lower end of the frequency band (0.01–0.2 Hz). Although the frequency dependence of the derived tβ* is quite simple, having smooth decreases from 0.01 to 5.0 Hz, a relatively complex parameterization of the frequency and depth dependence of Q is required to match this frequency dependence. Either the power n in a frequency‐dependent Q = Cfn varies with depth, the high‐frequency cutoff in the relaxation spectrum varies with depth, or both. The midmantle between depths of 400 and 1600 km contributes primarily to the attenuation of low frequencies (0.01–0.1 Hz) of body waves. Below the mantle depth of 2000 km the P and S waveforms for the paths investigated suggest that little or no attenuation exists in the band of 0.01–5 Hz.